Damage Quantification of Active Sensing Acousto-ultrasound-based SHM Based on a Multi-path Unit-cell Approach

Abstract

Assessing the reliability of damage quantification is a critical and necessary process for the evaluation of Nondestructive Evaluation (NDE) or Structural Health Monitoring (SHM) techniques. When it comes to NDE techniques, appropriate processes have been matured and established for asserting the reliability of damage quantification. However, such techniques offer solutions which that can be applied offline, are time consuming, and oftentimes quite expensive. On the other hand, for the case of SHM-based methods, and although several probabilistic methods have been proposed in the state-of-the-art literature, the task of damage quantification of active sensing techniques poses significant challenges that need to be properly addressed. Specifically, a major safety concern in aerospace structures is related to fatigue induced cracks for which accurate and reliable quantification is a critical issue for achieving the design performance and ensuring the aircraft safety. The main challenges associated with the quantification of fatigue cracks originate from the acousto-ultrasonic response discrepancies in similar structural materials that are due to variations in operating and environmental conditions, sensor positioning, damage characteristics (crack location, orientation, and propagation), and the effectiveness of the employed diagnostic algorithms. Recent studies have shown that apart from the operating/environmental conditions, the main sources of variation and uncertainty in the acoustic response are related to the location of the sensors and the damage location and propagation patterns. Trying to address the latter, this study presents an active sensing damage quantification approach that is based on the use of multiple piezoelectric sensors and corresponding acousto-ultrasonic damage propagation paths. Multiple paths from a multiple-sensor configuration unit, referred to as unit-cell, are used along with an adaptive weighted averaging method to mitigate the effects of sensor positioning errors and/or uncertainties associated with crack size and orientation. Several coupon-level experiments have been conducted to validate the performance of the method and investigate the convergence of the accuracy for increasing number of sensors.